Male-Specific Association Between MT-ND4 11719 A/G Polymorphism And

Dankowski et al. BMC Gastroenterology (2016) 16:118 DOI 10.1186/s12876-016-0509-1

RESEARCH ARTICLE Open Access Male-specific association between MT-ND4 11719 A/G polymorphism and ulcerative colitis: a mitochondria-wide genetic association study Theresa Dankowski1†, Torsten Schröder2,3†, Steffen Möller4,5, Xinhua Yu6,7, David Ellinghaus8, Florian Bär2, Klaus Fellermann2, Hendrik Lehnert2, Stefan Schreiber8, Andre Franke8, Christian Sina2, Saleh M. Ibrahim4* and Inke R. König1*

Abstract Background: Ulcerative colitis (UC) is a chronic inflammatory disorder of still unknown pathogenesis. Increasing evidence indicates that alterations in mitochondrial respiration and thus adenosine triphosphate (ATP) production are involved. This may contribute to mucosal energy deficiency and subsequently intestinal barrier malfunction, which is accepted to be a major hallmark of UC. Genetic alterations of the mitochondrial genome are one cause of mitochondrial dysfunction. However, less is known about mitochondrial gene polymorphisms in UC. Therefore, we aimed at identifying genetic associations between mitochondrial polymorphisms and UC. Methods: German UC cases (n = 1062) and German healthy controls (n = 3030) were genotyped using the Affymetrix Genome-Wide Human SNP Array 6.0. The primary association analysis was to test for associations between mitochondrial single nucleotide polymorphisms (SNPs) and UC using Fisher’s exact test in the total sample and stratified by sex. In addition, we tested for associations between mitochondrial haplogroups and UC and for interactions between the most promising mitochondrial SNPs and nuclear SNPs. An independent set of German subjects with 1625 UC cases and 3575 controls was used for replication. Results: We identified a genetic association between the MT-ND4 11719 A/G polymorphism and UC in the subgroup of males (rs2853495; odds ratio, 1.40; 95 % confidence interval, 1.13 to 1.73; p = 0.002). This association was replicated in the second independent cohort. In the association analysis based on mitochondrial haplogroups the lowest p values were reached for haplogroups HV and T (p = 0.029 and 0.035). Haplogroup HV is determined by the mitochondrial 11719 A/G polymorphism. Accordingly, this association was only found in the subgroup of males (p =0.009). (Continued on next page)

* Correspondence: [email protected]; [email protected] †Equal contributors 4The Lübeck Institute of Experimental Dermatology, University of Lübeck, Ratzeburger Allee 160, 23538 Lübeck, Germany 1Institute of Medical Biometry and Statistics, University of Lübeck, University Hospital Schleswig-Holstein, Campus Lübeck, Ratzeburger Allee 160, 23562 Lübeck, Germany Full list of author information is available at the end of the article

© 2016 The Author(s). Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Dankowski et al. BMC Gastroenterology (2016) 16:118 Page 2 of 7

(Continued from previous page) Conclusions: For the first time, we observed an association between the MT-ND4 11719 A/G polymorphism and UC. The gene MT-ND4 encodes for a subunit of the mitochondrial electron transport chain complex I, which is pivotal for ATP production and might therefore contribute to mucosal energy deficiency. The male-specific association indicates differences between males and females concerning the impact of mitochondrial gene polymorphisms on the development of UC. Further investigations of the functional mechanism underlying this association and the relevance of the gender-specificity are highly warranted. Keywords: Mitochondrial gene polymorphism, Ulcerative colitis, Male-specific association, Mucosal energy deficiency, Mitochondria-wide association study, MT-ND4, rs2853495, ATP, Haplogroup HV

Background We recently described that variants in mitochondrial Ulcerative colitis (UC) belongs to the family of inflam- DNA determine mucosal ATP levels and susceptibility matory bowel diseases (IBD). In western countries, UC to experimental colitis in mice [17]. In addition, in a exhibits rising prevalence with a slight predominance for small European UC cohort a mitochondrial polymorph- males [1]. UC is identified by chronic and recurrent in- ism was just recently described to be associated with UC testinal inflammation of yet unknown origin. The [18]. However, mitochondrial SNPs are usually excluded present pathophysiological concept suggests a disturbed from genetic associations studies, so we aimed at investi- intestinal barrier due to a combination of factors, includ- gating the role of mitochondrial gene variations in the ing dysregulation of the host’s mucosal immune system, development of UC. Here, we report on a mitochondria- environmental factors, changes in the intestinal micro- wide genetic association study using a large German UC biome and a genetic susceptibility [2–4]. Furthermore, study group and population-based controls. The restric- mucosal adenosine triphosphate (ATP) levels are re- tion to German subjects has the advantage that the Ger- duced in patients with UC [5, 6]. Since mitochondria man population is known to be genetically quite are the primary source of cellular ATP, this impli- homogenous [19]. As a result, a male-specific association cates a pathophysiological relevance of mitochondrial between a genetic variation of the MT-ND4 gene (11719 dysfunction. A/G, rs2853495) and UC could be identified and repli- Mitochondrial dysfunction is commonly caused by sin- cated in another independent set of German subjects. gle nucleotide polymorphism (SNPs) of the mitochondrial genome and inadequate repair mechanisms [7]. Methods Monogenic mutations of mitochondrial genes are known Study subjects to cause severe mitochondrial dysfunction leading to In total, 2687 German UC patients were available for rare and multisystemic diseases like Kearns Sayre Syn- this study who were recruited at the Department of drome, Leigh-syndrome or Leber’s Hereditary Optic General Internal Medicine of the Christian-Albrechts- Neuropathy [8–11]. According to the current know- University Kiel, the Charité University Hospital Berlin, ledge, these syndromes are not associated with UC. through local outpatient services, and nationwide with However, the mitochondrial DNA codes for mitochon- support of the German Crohn and Colitis Foundation. drial oxidative phosphorylation proteins and conse- Of these, 1043 German patients were previously used in quently may be of high relevance for the cellular energy a genome-wide association study (GWAS) for UC [20]. supply. Thus, cells with high need of energy to persevere Clinical, radiological, histological, and endoscopic (i.e. organ function are highly susceptible to mutations of the type and distribution of lesions) examinations were re- mitochondrial genome. Many cell types of the intestinal quired to unequivocally confirm the diagnosis of UC tract, such as epithelial cells, muscle cells, and immune [21]. Data from 1062 out of all 2687 UC cases composed cell, have high-energy requirements. Accordingly, it is of 459 males and 603 females were used in the initial highly likely that the mitochondrial DNA and respective study (more characteristics in Additional file 1: Table variants may be involved in gastrointestinal symptoms. S1). Data of the remaining 1625 UC cases composed of In fact, gastrointestinal symptoms of irritable bowel syn- 706 males, 914 females and 5 patients with missing sex drome could already been linked to mitochondrial poly- were used for replication. morphisms [12]. Furthermore, there is increasing For the initial study 1795 healthy controls were selected evidence for a role of mitochondrial gene polymor- from the KORA F4 survey [22], which is an independent phisms in the development of common diseases such as population-based sample of the general population living type 2 diabetes mellitus, neurodegenerative diseases and in the region of Augsburg, Southern Germany. In several different types of cancer [13–16]. addition, data from 1240 German control individuals was Dankowski et al. BMC Gastroenterology (2016) 16:118 Page 3 of 7

obtained from the Popgen biobank [23] for the initial corresponds to a local significance level of 0.05/(27⋅3) = − study resulting in a total of 3035 controls (1565 males, 6.2⋅10 04 for testing 27 SNPs in three groups. 1470 females). For the replication study, data from 3575 For secondary association analysis we reconstructed controls from Popgen biobank composed of 1701 males haplogroups. The haplogroup assignment was conducted and 1874 females were used. Of note, 1214 controls sub- by using the web application HaploGrep [24] that is jects from the Popgen biobank and 489 controls from the based on Phylotree [25]. According to the recommenda- KORA survey, respectively, had been part of the previous tions we only included haplogroup assignments with a GWAS [20]. quality score above 90 % as determined by HaploGrep. The only exception was haplogroup H2a2a, whose qual- Genotyping ity score was 0 %, because this is the haplogroup of the Single nucleotide polymorphism (SNP) genotyping in the reference sequence rCRS. This yielded haplogroup as- discovery panel was performed using the Affymetrix® signments with good quality for 1059 cases and 3007 Genome-Wide Human SNP Array 6.0 (for details see controls. Although HaploGrep gives very specific subha- [20]). Genotype calling was performed using the Birdseed plogroups, we only used the corresponding main hap- v2 algorithm implemented in Affymetrix Power Tools ver- logroups that are most common in Europe (H, I, J, K, T, sion 1.12.0. Prior to genotype calling, samples with a low U, V, W, X, HV, JT) for further analysis. We compared contrast-quality control value (contrast-qc < 0.40) were the relative frequencies of the most common European excluded. haplogroups in our sample to the frequencies given by Mitomap [26] and found good agreement (Additional Quality control file 2: Table S2). We only included European hap- For quality control on individual level in the initial study logroups with a frequency above 2 % and used Fisher’s group we excluded samples with a call fraction < 90 % exact test to test each haplogroup against all other hap- for the mitochondrial SNPs. This applied to five control logroups in the total initial study group and separate for individuals resulting in data for 1062 UC cases and 3030 males and females. Haplogroups H, V and HV were ex- controls. On mitochondrial marker level we excluded amined as one group named HV because of their close SNPs with MAF < 0.5 %. This applied to 60 of the initial relationship. Allele counts for all markers in the tested 119 SNPs. Nine SNPs were excluded because of a call haplogroups are provided in Additional file 3: Table S3. fraction < 90 % separate for cases and controls and two The functional relevance of the mitochondrial SNP SNPs because they were wrongly assigned to the mito- rs2853495 may be affected by the nuclear genome. Thus, chondrial DNA. Furthermore, we investigated the cluster to further explore our findings, we tested for an inter- plots of the remaining mitochondrial SNPs and excluded action between the most promising mitochondrial SNP 21 SNPs with critical issues. In conclusion, 27 mitochon- rs2853495 and nuclear SNPs in the total initial study drial SNPs passed our quality control. In Europeans 144 group and the subgroup of males. We used a logistic re- variants with a frequency >1 % were identified [13]. 37 gression model for each nuclear SNP that included the out of these 144 variants are captured or partly captured main effects of the nuclear and the mitochondrial SNPs by our 27 SNPs after quality control. and the interaction term. We modeled additive effects For quality control on nuclear marker level we ex- for the nuclear SNPs and we used a 0, 1 allele dosage cluded SNPs with MAF < 1 %, call fraction < 98 % separ- coding for the mitochondrial SNPs. A genome-wide sig- − − ate for cases and controls and p value < 10 04 for test on nificance level of 5⋅10 08 was used for the interaction Hardy Weinberg equilibrium in the controls. We also analysis. excluded SNPs on the X or Y chromosome. 546,808 of the initial 934,968 nuclear SNPs passed our quality Software control. We used the free software R [27] and Plink [28] for quality control and analysis. Statistical analysis Our primary association analysis in the initial study Results group was based on testing for associations between sin- Table 1 lists SNPs with a p value < 0.01 in the total initial gle SNPs and the UC phenotype by applying Fisher’s sample or in one of the sex-specific subgroups. Association exact test. This was done for the total initial study group plots for all SNPs in the total initial sample and both sex- and separate for males and females. The most promising specific subgroups are shown in Additional file 4: Figure S1, SNPs from this approach with p value < 0.01 were taken Additional file 5: Figure S2 and Additional file 6: Figure S3. forward to replication. To analyze joint effects across The association of the SNP rs2853495 in the gene MT-ND4 both stages, the Cochran Mantel Haenszel test was used. with UC was successfully validated in the total sample, and The global significance level is set to 0.05, which the combined analyses of this SNP across both stages Dankowski et al. BMC Gastroenterology (2016) 16:118 Page 4 of 7

Table 1 Results from the single SNP association analysis for SNPs with a two-sided p value < 0.01 from Fisher’s exact test in the initial sample or in one of the sex-specific subgroups mtSNPa BPb A1c A2d Subsample Initial analysis Replication Combinedg pORe A1c frequency p ORe A1 frequency p ORe Cases Ctrlsf Cases Ctrlsf rs2853495 11719 G A Total 0.007 1.22 [1.06; 1.40] 0.528 0.478 0.011 1.17 [1.04; 1.31] 0.525 0.486 1.921⋅10−4 1.19 [1.09; 1.30] Males 0.002 1.40 [1.13; 1.73] 0.562 0.479 0.002 1.32 [1.10; 1.57] 0.551 0.482 1.512⋅10−5 1.35 [1.18; 1.55] Females 0.328 1.10 [0.91; 1.34] 0.502 0.477 –– – – – – rs3899498 13368 A G Total 0.005 0.71 [0.56; 0.91] 0.085 0.115 0.334 0.90 [0.74; 1.10] 0.091 0.100 1.012⋅10−2 0.82 [0.70; 0.95] Males 0.126 0.75 [0.53; 1.07] 0.090 0.117 –– – – – – Females 0.026 0.68 [0.49; 0.96] 0.080 0.113 –– – – – – aMitochondrial single nucleotide polymorphism; bBase-pair position according to UCSC version hg19; cEffect allele; dSecond allele; eOdds ratio with 95 % confidence interval; fControls; gCombined analysis of initial and replication data

− yielded a p value of 1.9⋅10 4 (odds ratio (OR), 1.19; 95 % The results for the top 50 nuclear SNPs according to the p confidence interval (95 % CI), 1.09 to 1.30). This is signifi- value for interaction are shown in Additional file 7: Table cant on a global significance level of 0.05, which corre- S4. Additional file 8: Table S5 shows the results for all nu- − − sponds to a local significance level of 0.05/(27⋅3) = 6.2⋅10 4 clear SNPs with interaction p value < 1⋅10 04 in the sub- adjusted for testing of 27 mitochondrial SNPs in three group of males. groups. In the sex-specific analyses the association between rs2853495 and UC was only identified in the subgroup of Discussion males. This was also validated using the male subgroup of As the salient finding of this study, we identified an associ- the replication data. The combined analyses across both ation between a homoplastic mitochondrial DNA vari- − stages yielded a significant p value of 1.5⋅10 5 (OR, 1.35; ation in the gene MT-ND4 (11719 A/G, rs2853495) with 95 % CI, 1.18 to 1.55). Notably, the p value was smaller in male UC patients. This association was even validated in a thesubgroupofmaleseventhoughthesamplesizewas second independent sample, and the combined value is smallerthaninthetotalinitialsample. significant after controlling for the multiple testing of 27 The second approach was to test for association be- mitochondrial SNPs in three groups. tween mitochondrial haplogroups and UC (see Table 2). The here reported mtDNA variation (11719 A/G, The lowest p values in the total sample were reached for rs2853495) affects the gene MT-ND4, which encodes for haplogroups HV and T (p = 0.029 and 0.035), in which one subunit of the NADH dehydrogenase 4 of complex I haplogroup HV is determined by the mitochondrial of the mitochondrial respiratory chain. Since this endo- 11719 A/G polymorphism. Accordingly, the association symbiosis of mitochondria eukaryotic cells, the mito- to haplogroup HV was only found in the subgroup of chondria have lost most of their genetic material to the males (p = 0.009). cell nucleus. However, mitochondria still retained some No interaction between SNP rs2853495 and a nuclear genes in their own genome, which are highly essential − SNP reached genome-wide significance, i.e. p <5⋅10 8,in for cellular respiration and ATP generation [29]. The the total initial sample but some of the results merit fur- gene MT-ND4 encodes for one subunit of the NADH ther attention. Specifically, a cluster of SNPs on chromo- dehydrogenase 4 as a part of the mitochondrial oxidative some 3 shows interaction with rs2853495 (see Table 3). phosphorylation system (OXPHOS) complex I. Complex I consists of 44 different subunits, seven of which are mitochondrially encoded [30]. The function of complex Table 2 Results of the haplogroup based analysis with I as first step of the electron transport chain is to extract frequencies of cases and controls in the total sample and p values from Fisher’s exact test electrons from NADH and donate them to ubiquinone. This reaction releases energy, which is used to transport Haplogroup Control Cases pa p b p c males females protons across the mitochondrial inner membrane. In HV 48.05 52.03 0.029 0.009 0.498 this way, complex I contributes to the maintenance of U 15.96 15.30 0.624 0.056 0.322 the proton gradient, which fuels mitochondrial ATP pro- J 9.94 9.82 0.952 0.060 0.147 duction and many other mitochondrial functions [31]. T 11.64 9.25 0.035 0.357 0.060 Mutations in the MT-ND4 gene may have an important K 7.02 7.08 0.944 1.000 0.929 influence on mitochondrial respiratory chain function ap value in the total sample, bp value in the subgroup of males, cp value in the and therefore a genetic variation of this gene may lead subgroup of females to alterations of the cellular energy metabolism. Dankowski et al. BMC Gastroenterology (2016) 16:118 Page 5 of 7

Table 3 Results for the top 10 nuclear SNPs on chromosome 3 according to p value for interaction with rs2853495, sorted by chromosomal positions Chra Nuclear BPb A1c p Gene SNP Nucleard Mitoe Interactionf 3 rs7620175 35155767 T 6.58⋅10−02 4.25⋅10−01 8.47⋅10−06 LOC101928135 rs12493494 60412390 T 1.41⋅10−03 4.37⋅10−06 1.57⋅10−05 FHIT rs505014 62731917 G 2.12⋅10−04 1.11⋅10−01 5.78⋅10−05 CADPS rs498746 62731985 T 2.64⋅10−04 1.48⋅10−01 9.26⋅10−05 CADPS rs4676732 121314320 T 2.74⋅10−02 9.31⋅10−02 6.99⋅10−05 FBXO40 rs6784995 122938098 A 7.04⋅10−04 9.13⋅10−07 3.37⋅10−05 SEC22A rs9289215 122962809 A 2.72⋅10−03 1.37⋅10−06 8.36⋅10−05 SEC22A rs6784930 123001494 A 1.76⋅10−02 1.21⋅10−06 1.03⋅10−05 ADCY5 rs17809756 159630084 A 4.30⋅10−06 7.33⋅10−01 7.79⋅10−06 Near SCHIP1, ILI2A rs6800685 190139275 T 5.61⋅10−02 2.10⋅10−01 1.12⋅10−05 Near CLDN16, TMEM207 aChromosome; bBase-pair position according to UCSC version hg19; cMinor allele; dMain effect of nuclear SNP; eMain effect of mitochondrial SNP; fInteraction term

Consequently, a mutation in this mitochondrial gene highly energy-dependent pathways such as Calcium- may contribute among many other factors to mitochon- dependent secretion activator 1 (CADPS) and the drial dysfunction and mucosal energy deficiency, which Vesicle-trafficking protein SEC22A [40, 41]. However, has been detected in UC patients [5, 6]. more detailed data, e.g. provided by deep sequencing, The here identified DNA variation in MT-ND4 has are necessary to answer the question, whether there are been previously shown to be associated with numerous functionally active interactions between mitochondrial classical mitochondrial disorders like Leber’s Hereditary and nuclear gene variations. Optic Neuropathy [32] or Leigh syndrome [30]. More- Considering that homoplasmic mitochondrial DNA over, dysfunction of MT-ND4 has been described in the variants, like the here identified polymorphism, fre- context of experimental autoimmune encephalomyelitis, quently determine haplogroups [42], we tested for asso- which is an animal model of multiple sclerosis [33], ciations between different mitochondrial haplogroups breast cancer [34], cystic fibrosis [35] and other diseases. and UC. The lowest p values were reached for the mito- Most interestingly, a recent report in a small cohort of chondrial haplogroups HV and T (p = 0.029 and 0.035). UC patients could bring a different mitochondrial DNA Notably, the mitochondrial 11719 A/G polymorphism variant affecting MT-ND4 into relation to UC, which un- signifies the haplogroup HV [43]. Considering that these derlines the possible relevance for intestinal homeostasis haplogroups occur approximately in 40-50 % of modern [18]. Of note, we found the association only for male Eurasian ethnicity, in which the incidence of IBD is par- subjects. In fact, there are several studies reporting UC ticularly high [44, 45], these results might further point to be more frequent in males [1, 36, 37]. Moreover, it to a critical role of the mitochondrial genome for the was observed that males responded less to a three pathogenesis of UC. Haplogroup T is additionally de- months corticosteroidal therapy in terms of mucosal fined by a signature SNP of complex I [46, 47]. Hap- healing [38]. Beside these reports, it is well known that logroup T is currently found with high concentration in male mice respond aggravated to experimental colitis the eastern Baltic Sea region, in which the incidence of [39]. However, it remains speculation whether mito- IBD is emerging [45]. Whether this might be a func- chondrial gene polymorphisms may be implicated in tional connection of mitochondrial haplogroups and UC these clinical observations. Therefore, we alternatively or just a coincidence must be investigated in the future. suggest the presence of cofactors functionally cooperat- ing with the polymorphism of MT-ND4, which increase Conclusions the susceptibility for UC. These cofactors could for in- We identified and replicated a yet unknown male-specific stance be nuclear encoded genes. Accordingly, we stud- association between a mitochondrial gene polymorphism ied interactions between the identified mitochondrial in MT-ND4 (11719 A/G, rs2853495) and UC. This indi- SNP and nuclear SNPs. As apparent in Additional file 7: cates that mitochondrial genetics may determine gender- Table S4 and Additional file 8: Table S5 no genome-wide specific differences in disease prevalence and therapeutical significant interaction was found. Nevertheless, interest- response. Consequently, this study may help to deepen ing candidates emerged from these analyses. These SNPs the knowledge of UC pathology and clinical course. Fur- merit further investigation due to their involvement in ther studies are required to recapitulate the association of Dankowski et al. BMC Gastroenterology (2016) 16:118 Page 6 of 7

mitochondrial gene polymorphisms and UC and to eluci- Competing interests date the definite role of the mitochondrial genome in UC All authors declare no competing financial interest in relation to this work. All authors listed in the manuscript concur with the submission and none of development. the results have been previously reported or are under consideration for publication elsewhere. Additional files Consent for publication Not applicable. Additional file 1: Table S1. Characteristics of UC cases used in the initial analysis. (DOC 42 kb) Additional file 2: Table S2. Haplogroup frequencies in our sample and Ethics approval and consent to participate given by Mitomap (http://www.mitomap.org/bin/view.pl/MITOMAP/ Informed consent from all study participants was obtained. The study was HaplogroupMarkers, last accessed 15/09/2014). (DOC 43 kb) conducted in accordance with national and international laws and policies and was approved by the local ethics committee of the Christian-Albrechts- Additional file 3: Table S3. Allele counts in the tested haplogroups University of Kiel (156/03). All data was collected through studies as reported – assigned with HaploGrep (Kloss-Brandstätter 2011 Hum Mutat 32:25 32). earlier [20, 22, 23]. Base pair position and allele are shown for all markers. (DOC 107 kb) Additional file 4: Figure S1. Results of SNP based association analysis Author details in the total initial sample. Horizontal line located at 0.01. (DOC 33 kb) 1Institute of Medical Biometry and Statistics, University of Lübeck, University Additional file 5: Figure S2. Results of SNP based association analysis Hospital Schleswig-Holstein, Campus Lübeck, Ratzeburger Allee 160, 23562 Lübeck, Germany. 2Department of Medicine I, University of Lübeck, in the male subgroup of the initial sample. Horizontal line located at 0.01. 3 (DOC 34 kb) Ratzeburger Allee 160, 23538 Lübeck, Germany. Institute for Systemic Inflammation Research, University of Lübeck, Ratzeburger Allee 160, 23538 Additional file 6: Figure S3. Results of SNP based association analysis Lübeck, Germany. 4The Lübeck Institute of Experimental Dermatology, in the female subgroup of the initial sample. Horizontal line located at University of Lübeck, Ratzeburger Allee 160, 23538 Lübeck, Germany. 0.01. (DOC 33 kb) 5Institute for Biostatistics and Informatics in Medicine and Ageing Research, Additional file 7: Table S4. Results for the top 50 nuclear SNPs Rostock University Medical Center, Ernst-Heydemann-Straße 8, 18057 according to p value for interaction with rs2853495 in the total initial Rostock, Germany. 6Priority Area Asthma & Allergy, Research Center Borstel, sample, sorted by chromosomal positions. (DOC 86 kb) Airway Research Center North (ARCN), Members of the German Center for 7 Additional file 8: Table S5. Results for nuclear SNPs with p value < 1·10 Lung Research (DZL), Parkallee 1-40, 23845 Borstel, Germany. Xiamen-Borstel −04 Joint Laboratory of Autoimmunity, Medical College of Xiamen University, for interaction with rs2853495 in the subgroup of males, sorted by 8 chromosomal positions. (DOC 78 kb) Xiamen 361102, China. Institute of Clinical Molecular Biology, Christian-Albrechts-University of Kiel, Schittenhelmstraße 12, 24105 Kiel, Additional file 9: Data in aggregate form. (XLSX 16 kb) Germany.

Received: 22 March 2016 Accepted: 2 August 2016 Abbreviations ADP, adenosine diphosphate; ATP, adenosine triphosphate; GWAS, genome- wide association study; IBD, inflammatory bowel diseases; MAF, minor allele frequency; OXPHOS, oxidative phosphorylation; SNP, single-nucleotide poly- References morphism; UC, ulcerative colitis 1. Cosnes J, Gower-Rousseau C, Seksik P, Cortot A. Epidemiology and natural history of inflammatory bowel diseases. Gastroenterology. 2011;140(6):1785–94. Acknowledgements 2. Kaser A, Zeissig S, Blumberg RS. Inflammatory bowel disease. Annu Rev We thank all studied individuals with UC, their families and physicians for Immunol. 2010;28:573–621. their cooperation. We acknowledge the cooperation of the German Crohn 3. Rosenstiel P, Sina C, Franke A, Schreiber S. Towards a molecular risk map– and Colitis Foundation (Deutsche Morbus Crohn und Colitis Vereingung e.V.). recent advances on the etiology of inflammatory bowel disease. Semin We thank all individuals who functioned as healthy controls for this study. Immunol. 2009;21(6):334–45. 4. Gyires K, Toth EV, Zadori SZ. Gut inflammation: current update on pathophysiology, molecular mechanism and pharmacological treatment Funding modalities. Curr Pharm Des. 2014;20(7):1063–81. This study was supported by the BMBF through the National Genome 5. Sifroni KG, Damiani CR, Stoffel C, Cardoso MR, Ferreira GK, Jeremias IC, et al. Research Network (NGFN), by the Section of Medicine, University of Lübeck Mitochondrial respiratory chain in the colonic mucosal of patients with (P01-2014 to C.S.), the PopGen Biobank, and the Cooperative Research in the ulcerative colitis. Mol Cell Biochem. 2010;342(1–2):111–5. Region of Augsburg (KORA) research platform. KORA was initiated and 6. Santhanam S, Rajamanickam S, Motamarry A, Ramakrishna BS, Amirtharaj JG, financed by the Helmholtz Zentrum München – National Research Center Ramachandran A, et al. Mitochondrial electron transport chain complex for Environmental Health, which is funded by the German Federal Ministry of dysfunction in the colonic mucosa in ulcerative colitis. Inflamm Bowel Dis. Education, Science, Research and Technology and by the State of Bavaria 2012;18(11):2158–68. and the Munich Center of Health Sciences (MC Health) as part of the LMU 7. Wallace DC, Chalkia D. Mitochondrial DNA genetics and the heteroplasmy innovative initiative. The project received infrastructure support through the conundrum in evolution and disease. Cold Spring Harb Perspect Biol. 2013; DFG Clusters of Excellence “Inflammation at Interfaces”. 5(11):a021220. 8. Moraes CT, DiMauro S, Zeviani M, Lombes A, Shanske S, Miranda AF, et al. Availability of data and materials Mitochondrial DNA deletions in progressive external ophthalmoplegia and All data supporting our findings are available in aggregate form (Additional file 9). Kearns-Sayre syndrome. N Engl J Med. 1989;320(20):1293–9. 9. Riordan-Eva P, Harding AE. Leber’s hereditary optic neuropathy: the clinical Authors’ contributions relevance of different mitochondrial DNA mutations. J Med Genet. IRK, CS, and SMI conceived and designed the study. DE, SS and AF collected 1995;32(2):81–7. and prepared the data. TS, CS, KF, FB, HL, and SMI provided expertise in 10. Suzuki T, Nagao A, Suzuki T. Human mitochondrial tRNAs: biogenesis, inflammatory bowel diseases and mitochondrial function. TD, IRK, XY, and function, structural aspects, and diseases. Annu Rev Genet. 2011;45:299–329. SM performed the statistical analysis. TD, TS, IRK, CS and SMI interpreted the 11. Wang SB, Weng WC, Lee NC, Hwu WL, Fan PC, Lee WT. Mutation of results and drafted the manuscript. SM, XY, HL, KF, and FB participated in mitochondrial DNA G13513A presenting with Leigh syndrome, Wolff- critical review of the manuscript. All authors revised the manuscript critically Parkinson-White syndrome and cardiomyopathy. Pediatr Neonatol. for important intellectual content and approved the final manuscript. 2008;49(4):145–9. Dankowski et al. BMC Gastroenterology (2016) 16:118 Page 7 of 7

12. Wang WF, Li X, Guo MZ, Chen JD, Yang YS, Peng LH, et al. Mitochondrial 36. Molinie F, Gower-Rousseau C, Yzet T, Merle V, Grandbastien B, Marti R, et al. ATP 6 and 8 polymorphisms in irritable bowel syndrome with diarrhea. Opposite evolution in incidence of Crohn’s disease and ulcerative colitis in World J Gastroenterol. 2013;19(24):3847–53. Northern France (1988–1999). Gut. 2004;53(6):843–8. 13. Saxena R, de Bakker PI, Singer K, Mootha V, Burtt N, Hirschhorn JN, et al. 37. Gearry RB, Richardson A, Frampton CM, Collett JA, Burt MJ, Chapman BA, et Comprehensive association testing of common mitochondrial DNA al. High incidence of Crohn’s disease in Canterbury, New Zealand: results of variation in metabolic disease. Am J Hum Genet. 2006;79(1):54–61. an epidemiologic study. Inflamm Bowel Dis. 2006;12(10):936–43. 14. van der Walt JM, Nicodemus KK, Martin ER, Scott WK, Nance MA, Watts RL, 38. Ardizzone S, Cassinotti A, Duca P, Mazzali C, Penati C, Manes G, et al. et al. Mitochondrial polymorphisms significantly reduce the risk of Parkinson Mucosal healing predicts late outcomes after the first course of disease. Am J Hum Genet. 2003;72(4):804–11. corticosteroids for newly diagnosed ulcerative colitis. Clin Gastroenterol 15. Brandon M, Baldi P, Wallace DC. Mitochondrial mutations in cancer. Hepatol. 2011;9(6):483–9. e3. Oncogene. 2006;25(34):4647–62. 39. Babickova J, Tothova L, Lengyelova E, Bartonova A, Hodosy J, Gardlik R, et 16. Lakatos A, Derbeneva O, Younes D, Keator D, Bakken T, Lvova M, et al. al. Sex differences in experimentally induced colitis in mice: a role for Association between mitochondrial DNA variations and Alzheimer’s disease estrogens. Inflammation. 2015;38(5):1996–2006. in the ADNI cohort. Neurobiol Aging. 2010;31(8):1355–63. 40. Wassenberg JJ, Martin TF. Role of CAPS in dense-core vesicle exocytosis. 17. Bär F, Bochmann W, Widok A, von Medem K, Pagel R, Hirose M, et al. Ann N Y Acad Sci. 2002;971:201–9. Mitochondrial gene polymorphisms that protect mice from colitis. 41. Daily NJ, Boswell KL, James DJ, Martin TF. Novel interactions of CAPS (Ca2 + Gastroenterology. 2013;145(5):1055–63. e3. −dependent activator protein for secretion) with the three neuronal SNARE 18. Rosa A, Abrantes P, Sousa I, Francisco V, Santos P, Francisco D, et al. proteins required for vesicle fusion. J Biol Chem. 2010;285(46):35320–9. Ulcerative colitis is under dual (Mitochondrial and Nuclear) genetic control. 42. Herrnstadt C, Howell N. An evolutionary perspective on pathogenic mtDNA Inflamm Bowel Dis. 2016;22(4):774–81. mutations: haplogroup associations of clinical disorders. Mitochondrion. 19. Steffens M, Lamina C, Illig T, Bettecken T, Vogler R, Entz P, et al. SNP-based 2004;4(5–6):791–8. analysis of genetic substructure in the German population. Hum Hered. 43. Saillard J, Magalhaes PJ, Schwartz M, Rosenberg T, Norby S. Mitochondrial 2006;62(1):20–9. DNA variant 11719G is a marker for the mtDNA haplogroup cluster HV. 20. Franke A, Balschun T, Sina C, Ellinghaus D, Hasler R, Mayr G, et al. Genome- Hum Biol. 2000;72(6):1065–8. wide association study for ulcerative colitis identifies risk loci at 7q22 and 44. Brandstätter A, Salas A, Niederstätter H, Gassner C, Carracedo A, Parson W. 22q13 (IL17REL). Nat Genet. 2010;42(4):292–4. Dissection of mitochondrial superhaplogroup H using coding region SNPs. – 21. Lennard-Jones JE. Classification of inflammatory bowel disease. Scand J Electrophoresis. 2006;27(13):2541 50. Gastroenterol Suppl. 1989;170:2–6. discussion 16–9. 45. Burisch J, Munkholm P. Inflammatory bowel disease epidemiology. Curr – 22. Wichmann HE, Gieger C, Illig T, for the MONICA/KORA Study Group. KORA- Opin Gastroenterol. 2013;29(4):357 62. gen–resource for population genetics, controls and a broad spectrum of 46. SanGiovanni JP, Arking DE, Iyengar SK, Elashoff M, Clemons TE, Reed GF, et disease phenotypes. Gesundheitswesen. 2005;67 Suppl 1:S26–30. al. Mitochondrial DNA variants of respiratory complex I that uniquely 23. Krawczak M, Nikolaus S, von Eberstein H, Croucher PJ, El Mokhtari NE, characterize haplogroup T2 are associated with increased risk of age-related Schreiber S. PopGen: population-based recruitment of patients and controls macular degeneration. PLoS One. 2009;4(5):e5508. for the analysis of complex genotype-phenotype relationships. Community 47. Bedford FL. Sephardic signature in haplogroup T mitochondrial DNA. Eur J – Genet. 2006;9(1):55–61. Hum Genet. 2012;20(4):441 8. 24. Kloss-Brandstätter A, Pacher D, Schönherr S, Weissensteiner H, Binna R, Specht G, et al. HaploGrep: a fast and reliable algorithm for automatic classification of mitochondrial DNA haplogroups. Hum Mutat. 2011;32(1):25–32. 25. van Oven M, Kayser M. Updated comprehensive phylogenetic tree of global human mitochondrial DNA variation. Hum Mutat. 2009;30(2):E386–94. 26. MITOMAP. A human mitochondrial genome database. http://www.mitomap. org. Accessed 15 Sept 2014. 27. Core Team R. R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing; 2014. http://www.R-project.org. Accessed 15 Sept 2014. 28. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81(3):559–75. 29. Johnston IG, Williams BP. Evolutionary inference across eukaryotes identifies specific pressures favoring mitochondrial gene retention. Cell Syst. 2016;2(2):101–11. 30. Breuer ME, Willems PH, Smeitink JA, Koopman WJ, Nooteboom M. Cellular and animal models for mitochondrial complex I deficiency: a focus on the NDUFS4 subunit. IUBMB Life. 2013;65(3):202–8. 31. Koopman WJ, Nijtmans LG, Dieteren CE, Roestenberg P, Valsecchi F, Smeitink JA, et al. Mammalian mitochondrial complex I: biogenesis, regulation, and reactive oxygen species generation. Antioxid Redox Signal. 2010;12(12):1431–70. Submit your next manuscript to BioMed Central 32. Wallace DC, Singh G, Lott MT, Hodge JA, Schurr TG, Lezza AM, et al. and we will help you at every step: Mitochondrial DNA mutation associated with Leber’s hereditary optic neuropathy. Science. 1988;242(4884):1427–30. • We accept pre-submission inquiries 33. TallaV,YuH,ChouTH,PorciattiV,ChiodoV,BoyeSL,etal.NADH- • Our selector tool helps you to ﬁnd the most relevant journal dehydrogenase type-2 suppresses irreversible visual loss and neurodegeneration • We provide round the clock customer support in the EAE animal model of MS. Mol Ther. 2013;21(10):1876–88. 34. Santidrian AF, Matsuno-Yagi A, Ritland M, Seo BB, LeBoeuf SE, Gay LJ, et al. • Convenient online submission Mitochondrial complex I activity and NAD+/NADH balance regulate breast • Thorough peer review cancer progression. J Clin Invest. 2013;123(3):1068–81. • Inclusion in PubMed and all major indexing services 35. Valdivieso AG, Clauzure M, Marin MC, Taminelli GL, Massip Copiz MM, Sanchez F, et al. The mitochondrial complex I activity is reduced in cells • Maximum visibility for your research with impaired cystic fibrosis transmembrane conductance regulator (CFTR) function. PLoS One. 2012;7(11):e48059. Submit your manuscript at www.biomedcentral.com/submit